We are entering a new era in breast cancer research in which testing of interventional therapies to prevent the disease will be a major focus. Epidemiological studies have identified certain benign proliferative lesions associated with a modestly elevated risk of developing invasive breast cancer. It is likely that a subset of these lesions also acquire biological abnormalities which greatly increase their risk of evolving to breast cancer. In fact, preliminary results from Project 4 of this SPORE strongly suggest that "hyperplastic" breast lesions may indeed be direct clonal precursors of invasive breast cancer. Drugs which suppress breast cancer may also induce changes in these histological and biological markers of increased breast cancer risk, and measuring these changes may therefore be useful as surrogate endpoints in preliminary chemoprevention trials. This project is a bold but reasonable step to begin addressing these issues. Towards this end we propose to test the ability of tamoxifen to: (1) cause histological regression of premalignant hyperplastic breast lesions; and (2) revert towards normal the abnormal phenotypes of certain biomarkers associated with these premalignant lesions, evaluated by immunohistochemistry. We will randomize 100 women with biopsy-proven hyperplastic breast disease to either tamoxifen or no therapy for one year, followed by a second biopsy to assess whether or not either the morphology or the biomarkers of the lesions have improved. Biomarkers to be assessed are being chosen based on our ongoing study of receptors, growth factors, oncogenes, and proliferation markers in putative precursor lesions of invasive breast cancer. After considerable effort, this trial has recently received FDA approval, and the first several patients have already been biopsied, enrolled, and started on treatment. Success will provide preliminary support for these biomarkers as legitimate surrogate endpoints for definitive breast cancer chemoprevention trials.